STEM System and other VR Controls.

[mcilwham]

Hello Everyone,

I'm a little overwhelmed at all the technology coming out in kickstarter, and would like to voice my questions and concerns to a wider audience of enthusiasts and experts.

As most of you know, the STEM system is an up to 5 point tracking system using LMF (low magnetic frequencies). This is the same technology as in the Razer Hydra. The STEM System is sold as a low end motion capture suit, that allows for arm, leg and chest/head tracking. Theoretically, the STEM would be a great tool and a great interface for VR and the Oculus Rift. Who wouldn't want to be able to swing a sword or aim a pistol as they would in real life?

I sure as hell would, that's why I did some more research on the topic. Lacking in videos that display actual game footage, the only way to give any idea on how the STEM will interact with the game world at large is the Razer Hydra. In current shooting games using the hydra, you don't aim using them. You aim using the sticks included on each hand piece.

https://www.youtube.com/watch?v=7yBbDjY95GY (Kinda long, only need like 2 minutes to get the point)

Skyrim at least has the ability to aim with the damn thing, but still only uses motion for menus and gimics:
https://www.youtube.com/watch?v=7T1Rv7jrg5M

The STEM system is a massive improvement over the Razer Hydra, do not get it twisted. Its range is more then doubled, and the lack of wires makes the whole much better. The issue becomes that it could be wireless and have 10000 ft of range, if they only use the buttons and sticks its no better then a gamepad someone split in half. Does anyone know what plans they have for motion controls moving forward? Sixense claims they plan to have EVERY GAME work with the system, but what the hell is point if the motion doesn't factor in? Besides the hand tracking and control, this system only provides one thing: 360 Degree positional head tracking. For that, cant i just buy one of their bases and a single pack saving myself roughly 200$?

For the most part, all arm/weapon tracking seems incredibly flawed. Most games cant free aim (Light guns, Trinity Magnum), and less then no games have the ability to control finger movement (for the other option of ControlVR etc.) making the Head/Leg tracking the only viable options at the moment. With a STEM/Hydra/TrinityMagnum you should theoretically aim based on the weapon itself, but with almost every game I can think of railing the head to the back of the gun the Oculus itself, and whatever you are using to track it (be it the camera or the STEM etc.), will be what is used to aim the gun.

Maybe I'm nit picking, maybe I shouldn't expect games already made to have any sort or motion controls added as a secondary thought. All I know is the waiting for my SDK 2 is killing me, and all these other products sure do look great.

Thanks for your time, and sorry for the block of text.

(If i am wrong about anything, please feel free to correct me and I apologize in advance)

[saviornt]

Not entirely sure what you're asking.

If it is about support for "VR Controls", then it will need to be added in by the developers to be done correctly. Everything regarding the control system will need to be "upgraded" and proper animations / motion capture added. While its not entirely difficult, it does require time.

As it is now, you press button A, the game plays animation A; with a motion capture controller such as STEM/PrioVR/ControlVR, I would presume there would be sort of "gesture controls" to perform an action. Its up to the imagination on how to support it so that it isn't a PITA to use.

[mcilwham]

Thanks for the response.

The whole input A = Button A is what I assumed they would be using to make the tools backward compatible with games not designed/not being released with the VR technology available or in mind. Based on the Hydra though, they don't really do that. Perhaps hoping for someone who had been to a convention and tried them out or a similar experience to chime in will work.

[saviornt]

I've worked a bit on control schemes for Unity and UDK. You have to add support for the controls into the game using the appropriate API.

You can also use a 3rd party solution to basically emulate a control scheme, however, it won't be a proper implementation.

[owenwp]

What you are seeing there is Motioncreator, which lets you map different gestures and movements to buttons and mouse axis. It is only a tool that lets existing games work with the hardware, so that they are not completely unplayable.

Our SixenseVR SDK is completely separate, and provides all the tools you need to get exact 1:1 aiming.

[davidjc]

I am more excited for PrioVR at the moment. But only time will tell which system gets it right in the end.

[kojack]

"mcilwham" wrote:
Lacking in videos that display actual game footage, the only way to give any idea on how the STEM will interact with the game world at large is the Razer Hydra. In current shooting games using the hydra, you don't aim using them. You aim using the sticks included on each hand piece.

Have a look here:



In the 30s to 1min range it shows a Stem being used for full control of a gun in a game style environment.

For the Hydra, games that don't support it natively aren't going to show it off very well. Take a look at Portal 2, it has custom content just for the Hydra and it uses the Hydra's abilities much better. Motion controls not only aim the portal gun but also allow manipulation of objects in the world (orientation and position).

The same situation happens for any niche input device with unique features. 99.999999% of games are kind of pointless on a Novint Falcon, but the few that understand it are amazing (HalfLife 2 has every gun's recoil and reload modelled as forces, Penumbra allows you to feel individual pixels of every surface (it has 1000Hz force updates)).

There's some gun use in this PrioVR video:

[mcilwham]

Thank you all for the replies, as spending 400$ is always best well thought out.

I have seen that video of the weapon use, which is why I am excited and considering the purchase.It's potential is amazing, but I was worried by lack of creativity/use of that potential. I have seen the Novint before as well, very neat device. I guess you are right that because it is a niche tool it will have a niche game group well still working with the general public. The controls look good to me, the hydra was more then interesting, and the added 360 degree full positional tracking for the head is more then icing on the cake.

All I can hope is it gains the community support for mods etc that the rift has gained.

[nuB]

"davidjc" wrote:
I am more excited for PrioVR at the moment. But only time will tell which system gets it right in the end.

A good approach to evaluating new VR control peripherals is taking note of possible drift, the device accuracy, and latency.

This is why I am getting a STEM system and not priovr. Due to drift, latency, and accuracy. I need them as close to 1:1 as possible, and priovr does not meet my quality standards.


APIs in popular engines should allow for easy adoption of these control systems.

[davidjc]

"nuB" wrote:

"davidjc" wrote:
I am more excited for PrioVR at the moment. But only time will tell which system gets it right in the end.

A good approach to evaluating new VR control peripherals is taking note of possible drift, the device accuracy, and latency.

This is why I am getting a STEM system and not priovr. Due to drift, latency, and accuracy. I need them as close to 1:1 as possible, and priovr does not meet my quality standards.


APIs in popular engines should allow for easy adoption of these control systems.

Everything I have heard mentioned that PrioVR was more accurate as far as latency, drift I don't know which will be better. I do know that it was mentioned that PrioVR might be behind STEM as far as moving about the room in the suit. As it depends on caclulating where you should be based on skeletal movements tracked by the ProiVR sensors.

So this is how it was a few months ago. I am sure both systems are trying to shore up their weaknesses so it will be interesting to see how it plays out. STEM is much more expensive as well.